Loading NHSRdatasets
Mairead Bermingham
27 May, 2022
Overview
In this lesson we will load the five datasets from the NHSRdatasets package. Next we will briefly view, explore and tabulate the NHS England accident and emergency attendances and admissions (ae_attendances) data set and save it to your ‘RawData’ folder. We will then examine the four-hour waiting time target performance for England as a whole, selecting a subset of the variables needed. We will then partitioned the data subset into training and testing data and save them to your working ‘Data’ folder, ready for downstream exploratory analysis.
Background
The data that you will be managing on the course are from the NHSRdatasets package. This package has been created to support skills development in the NHS-R community. It contains several free datasets, including:
NHS England accident and emergency attendances and admissions (ae_attendances). Reported attendances, four-hour breaches and admissions for all A&E departments in England for the years 2016/17 through 2018/19 (Apr-Mar).
Hospital length of stay (LOS) data (LOS_model). Artificially generated hospital data. Fictional patients at ten fictional hospitals, with LOS, age and date status data. Data were generated to learn generalised linear models (GLM) concepts, modelling either death or LOS.
Deaths registered weekly in England and Wales, provisional (ons_mortality). Provisional counts of the number of deaths registered in England and Wales, by age, sex and region, in the latest weeks for which data are available.
Stranded patient (Patients flagged as having a greater than 7 day LOS) model (stranded_data). This model is to be used as a machine learning classification model for supervised learning. The binary outcome is stranded vs not stranded patients.
Synthetic national early warning scores data (synthetic_news_data). Synthetic NEWS data to show the results of the NHSR_synpop package. NEWS is a tool developed by the Royal College of Physicians which improves the detection and response to clinical deterioration in adult patients and is a key element of patient safety and improving patient outcomes. NEWS2 has been endorsed by NHS England and NHS Improvement – for use in acute and ambulance settings. These datasets have been synthetically generated. (Chris Mainey, 2021)
Load packages and data
Let’s load the packages and data needed for this script.
library(NHSRdatasets)
library(tidyverse)
Illustration by Allison Horst
The tidyverse is a collection of R packages designed for data science introduced by Hadley Wickham and his team that “share an underlying design philosophy, grammar, and data structures” of tidy data.
library(here)The here package enables easy file referencing in project-oriented workflows. In contrast to using setwd() function, which is fragile and dependent on the way you organise your files, here uses the top-level directory of a project to easily build paths to files.
library(knitr)knitr is an R package that integrates code into text documents. Files can then be processed into a diverse array of document formats including pdfs, Word documents, etc.
library(scales)The scales packages provides the internal scaling infrastructure to ggplot2 and its functions allow programmers to customise the transformations, breaks, guides and palettes used in visualisations. ggplot2 is a powerful package to draw graphics. It implements the grammar of graphics (and hence its name). ggplot2 is included in, and loaded with the tidyverse package
library(lubridate)lubridate provides tools that make it easier to manipulate dates in R. lubridate is part of Hadley’s tidyverse ecosystem but is not loaded by the tidyverse package, which includes only what he thought were the core components. If you are going to be doing a lot of date manipulations, you need to load it separately.
library(caret)The caret package (short for Classification And REgression Training) is a set of functions that attempt to streamline the process for creating predictive models. The package contains tools for data splitting, pre-processing, feature selection, model tuning using resampling, variable importance estimation amongst others.
NHSRdatasets
Here is the code to load the five NHSRdatasets from the NHSRdatasets R package.
#Load the ae_attendances data.
data(ae_attendances)
#Load the LOS_model data.
data(LOS_model)
#Load the ons_mortality data.
data(ons_mortality)
#Load the stranded_data.
data(stranded_data)
#Load the synthetic_news_data.
data(synthetic_news_data)NHS England accident and emergency attendances and admissions (ae_attendances) data
In this course, we will be focus on the NHS England accident and emergency attendances and admissions (ae_attendances) data. If you are new to programming, I recommend using this data, as the data has been tidied to be easily usable within the tidyverse of packages. However, if you would like to stretch yourself for the course assignment, I recommend you can choose from one of the other four datasets contained within the NHSRdatasets package to manage on this course. You will not gain additional marks, but it will significantly enhance your learning on the course.
Let’s have a look at the ae_attendances data
data(ae_attendances)
ae<-ae_attendances
class(ae)## [1] "tbl_df" "tbl" "data.frame"The tbl_df class is a subclass of data.frame. A data.frame is a table or a two-dimensional array-like structure in which each column contains values of one variable and each row contains one set of values from each column. The colloquial term “tibble” refers to a data frame that has the tbl_df class. Objects of class tbl_df have a class attribute of c(“tbl_df”, “tbl”, “data.frame”). Tibble is the central data structure for the set of packages known as the tidyverse. It’s best to think of tibbles as an updated and stylish version of the data.frame that exhibit some special behaviour, such as enhanced printing.
ae## # A tibble: 12,765 × 6
## period org_code type attendances breaches admissions
## <date> <fct> <fct> <dbl> <dbl> <dbl>
## 1 2017-03-01 RF4 1 21289 2879 5060
## 2 2017-03-01 RF4 2 813 22 0
## 3 2017-03-01 RF4 other 2850 6 0
## 4 2017-03-01 R1H 1 30210 5902 6943
## 5 2017-03-01 R1H 2 807 11 0
## 6 2017-03-01 R1H other 11352 136 0
## 7 2017-03-01 AD913 other 4381 2 0
## 8 2017-03-01 RYX other 19562 258 0
## 9 2017-03-01 RQM 1 17414 2030 3597
## 10 2017-03-01 RQM other 7817 86 0
## # … with 12,755 more rowsAn overview of your data is returned when a tibble is printed to the RStudio console. You can see the ae_attendances tibble consists of 12,765 rows of data and six columns with different classes. You have one date variable, period, two character variables (or factors), org_code and type, and three numeric (double precision) variables, attendances, breaches and admissions. A factor is a variable used to categorise and store the data, having a limited number of different values. Factors are an important class for exploratory data analysis, visualisation, and statistical analysis with R.
The dataset contains:
period: the month that this activity relates to, stored as a date (1st of each month).
org_code: the Organisation data service (ODS) code for the organisation. The ODS code is a unique code created by the Organisation data service within NHS Digital, and used to identify organisations across health and social care. ODS codes are required in order to gain access to national systems like NHSmail and the Data Security and Protection Toolkit. If you want to know the organisation associated with a particular ODS code, you can look it up from the following address: https://odsportal.digital.nhs.uk/Organisation/Search. For example, the organisation associated with the ODS code ‘AF003’ is Parkway health centre.
type: the Department Type for this activity, either
- 1: Emergency departments are a consultant led 24 hour service with full resuscitation facilities and designated accommodation for the reception of accident and emergency patients,
- 2: Consultant led mono specialty accident and emergency service (e.g. ophthalmology, dental) with designated accommodation for the reception of patients, or
- other: Other type of A&E/minor injury activity with designated accommodation for the reception of accident and emergency patients. The department may be doctor led or nurse led and treats at least minor injuries and illnesses and can be routinely accessed without appointment. A service mainly or entirely appointment based (for example a GP Practice or Out-patient clinic) is excluded even though it may treat a number of patients with minor illness or injury. Excludes NHS walk-in centres.(National Health Service, 2020)
- 1: Emergency departments are a consultant led 24 hour service with full resuscitation facilities and designated accommodation for the reception of accident and emergency patients,
attendances: the number of attendances for this department type at this organisation for this month.
breaches: the number of attendances that breached the four hour target.
admissions: the number of attendances that resulted in an admission to the hospital.(Chris Mainey, 2021)
Let’s view at the ae_attendances data
The glimpse() function is from tibble package and is great to view the columns/variables in a data frame. It also shows data type and some of the data in the data frame in each row. The tibble package provides utilities for handling tibbles. The tibble package is loaded by the tidyverse package, as one of its core components.
glimpse(ae) ## Rows: 12,765
## Columns: 6
## $ period <date> 2017-03-01, 2017-03-01, 2017-03-01, 2017-03-01, 2017-03-0…
## $ org_code <fct> RF4, RF4, RF4, R1H, R1H, R1H, AD913, RYX, RQM, RQM, RJ6, R…
## $ type <fct> 1, 2, other, 1, 2, other, other, other, 1, other, 1, other…
## $ attendances <dbl> 21289, 813, 2850, 30210, 807, 11352, 4381, 19562, 17414, 7…
## $ breaches <dbl> 2879, 22, 6, 5902, 11, 136, 2, 258, 2030, 86, 1322, 140, 0…
## $ admissions <dbl> 5060, 0, 0, 6943, 0, 0, 0, 0, 3597, 0, 2202, 0, 0, 0, 3360…Here is the output of the `glimpse()’ function. It starts off with the number of rows and columns and each column in separate rows.
The head() function let’s you get a look at top n rows of a data frame. By default it shows the first 6 rows in a data frame.
head(ae)## # A tibble: 6 × 6
## period org_code type attendances breaches admissions
## <date> <fct> <fct> <dbl> <dbl> <dbl>
## 1 2017-03-01 RF4 1 21289 2879 5060
## 2 2017-03-01 RF4 2 813 22 0
## 3 2017-03-01 RF4 other 2850 6 0
## 4 2017-03-01 R1H 1 30210 5902 6943
## 5 2017-03-01 R1H 2 807 11 0
## 6 2017-03-01 R1H other 11352 136 0We can specify the number of rows we want to see in a data frame with the argument “n”.
head(ae, n=3)## # A tibble: 3 × 6
## period org_code type attendances breaches admissions
## <date> <fct> <fct> <dbl> <dbl> <dbl>
## 1 2017-03-01 RF4 1 21289 2879 5060
## 2 2017-03-01 RF4 2 813 22 0
## 3 2017-03-01 RF4 other 2850 6 0In the example above, we used n=3 to look at the first three rows of the ae_attendances data frame.
The function tail() is the counterpart to head(). tail() let’s you to take a look at the bottom n rows of a data frame. We can adjust the number of rows with the argument “n” as with the head() function.
tail(ae, n=4)## # A tibble: 4 × 6
## period org_code type attendances breaches admissions
## <date> <fct> <fct> <dbl> <dbl> <dbl>
## 1 2018-04-01 RA3 1 3825 476 1016
## 2 2018-04-01 AXG other 2980 24 0
## 3 2018-04-01 NLX24 other 1538 0 0
## 4 2018-04-01 RA4 1 4388 82 1292In the example above, we used n=4 to look at the last four rows of the ae_attendances data frame.
Missing data
It is common when collecting data for some entries to be missing, which can significantly impact any attempt to gain useful insight from data. Methods have been developed to handle missing data. The simplest of these is to discard any record which contains a missing entry. However, this can lead to such a small sample that it is not useful for obtaining reliable information. In addition to this, there may be reasons why certain demographic groups do not want to supply particular information, reducing the representativeness of the selected sample. We therefore need to check for missing data in the ae_attendances data.
Missing data (source: Matt Randall, 2020)
‘map()’ is a function from the purr package for applying a function to each element of a list. The purr package enhances R’s functional programming toolkit by providing a complete and consistent set of tools for working with functions and vectors. The ‘is.na’ function produces a matrix, consisting of logical values (i.e. TRUE or FALSE), whereby TRUE indicates a missing value.
# Calculate how many NAs there are in each variable
ae %>%
map(is.na) %>%
map(sum)## $period
## [1] 0
##
## $org_code
## [1] 0
##
## $type
## [1] 0
##
## $attendances
## [1] 0
##
## $breaches
## [1] 0
##
## $admissions
## [1] 0The data is complete. We do not need to worry about handling missing data. If you decide use another NHSRdataset it may not be as ‘tidy’ as the ae_attendances data.
Let’s add an index link column to ae_attendances data
Separating data into training and testing sets is vital for evaluating data collection and analysis tools. Typically, when you divide a data set into a training set and a testing set, most of the data is used for training, and a smaller portion of the data is used for testing. To develop and evaluate your data capture tool, you will need to split the raw data into training and testing sets. We, therefore, need to add an index column to the raw data so we can link the partitioned data sets to the raw data if required in the future. We will use the rowid_to_column() function to convert row identities to a column named index.
ae <- rowid_to_column(ae, "index")Let’s tablulate the raw data for your report
The pipe operator denoted by symbol %>% is a special operational function available which allows us to pass the result of one function/argument to the other one in sequence. The the dplyr package mutate_at() function edits specific columns with a character vector or ‘vars()’ argument generates a list of columns. The function comma() from the scales package is useful for presenting numbers using commas to separate the thousands. This is always a good idea as it assists the reader in quickly determining the magnitude of the numbers we are looking at. As a matter of course, I recommend commas in plots (and tables) at all times.
ae %>%
# Set the period column to show in month-year format
mutate_at(vars(period), format, "%b-%y") %>%
# Set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches, admissions), comma) %>%
# Show the first 10 rows
head(10) %>%
# Format as a table
kable()| index | period | org_code | type | attendances | breaches | admissions |
|---|---|---|---|---|---|---|
| 1 | Mar-17 | RF4 | 1 | 21,289.0 | 2,879.0 | 5,060.0 |
| 2 | Mar-17 | RF4 | 2 | 813.0 | 22.0 | 0.0 |
| 3 | Mar-17 | RF4 | other | 2,850.0 | 6.0 | 0.0 |
| 4 | Mar-17 | R1H | 1 | 30,210.0 | 5,902.0 | 6,943.0 |
| 5 | Mar-17 | R1H | 2 | 807.0 | 11.0 | 0.0 |
| 6 | Mar-17 | R1H | other | 11,352.0 | 136.0 | 0.0 |
| 7 | Mar-17 | AD913 | other | 4,381.0 | 2.0 | 0.0 |
| 8 | Mar-17 | RYX | other | 19,562.0 | 258.0 | 0.0 |
| 9 | Mar-17 | RQM | 1 | 17,414.0 | 2,030.0 | 3,597.0 |
| 10 | Mar-17 | RQM | other | 7,817.0 | 86.0 | 0.0 |
Let’s save the raw ae_attendances data to your ‘RawData’ folder
We will use the here package to build a path relative to the top-level directory to write the raw ae_attendances data to your ‘RawData’ folder. The goal of the here package is to enable easy file referencing in project-oriented workflows. In contrast to using setwd() function, which is fragile and dependent on the way you organise your files, here uses the top-level directory of a project to easily build paths to files. 
write_csv(ae, here("RawData", "ae_attendances.csv"))Selecting variables for your data capture tool
It really important point to note, is that you are not going to need the full ae_attendances dataset develop your data collection tool. For example if you are interested in exploring four hours performance for England as a whole you will only need the variables: index, period, attendances, breaches. Likewise, you may wish examine performance for the different types of department, so we would also need the type variable. For the purpose of this lesson, we will focus on England as a whole.
Examining the four-hour waiting time target performance for England
We will now use the dplyr package select() function to select the required variables. The dplyr package is loaded by the tidyverse package, as one of its core components. dplyr provides a grammar of data manipulation, providing a consistent set of verbs that solve the most common data manipulation challenges.
ae<-ae %>% select(index, period, attendances, breaches)Let’s tabulate the subsetted ae_attendances data
ae %>%
# set the period column to show in Month-Year format
mutate_at(vars(period), format, "%b-%y") %>%
# set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches), comma) %>%
# show the first 10 rows
head(10) %>%
# format as a table
kable()| index | period | attendances | breaches |
|---|---|---|---|
| 1 | Mar-17 | 21,289.0 | 2,879.0 |
| 2 | Mar-17 | 813.0 | 22.0 |
| 3 | Mar-17 | 2,850.0 | 6.0 |
| 4 | Mar-17 | 30,210.0 | 5,902.0 |
| 5 | Mar-17 | 807.0 | 11.0 |
| 6 | Mar-17 | 11,352.0 | 136.0 |
| 7 | Mar-17 | 4,381.0 | 2.0 |
| 8 | Mar-17 | 19,562.0 | 258.0 |
| 9 | Mar-17 | 17,414.0 | 2,030.0 |
| 10 | Mar-17 | 7,817.0 | 86.0 |
Let’s calculate monthly four hour waiting time target performance for England as a whole
We will now use the dplyr package summarise_at() function to get the sum of all attendances and breaches for England as a whole.
ENG_performance <- ae %>%
group_by(period) %>%
summarise_at(vars(attendances, breaches), sum) %>%
mutate(performance = 1 - breaches / attendances)
glimpse(ENG_performance)## Rows: 36
## Columns: 4
## $ period <date> 2016-04-01, 2016-05-01, 2016-06-01, 2016-07-01, 2016-08-0…
## $ attendances <dbl> 1867781, 2070340, 1958802, 2079034, 1932901, 1952464, 2001…
## $ breaches <dbl> 186122, 201329, 184912, 201973, 174419, 182597, 219137, 22…
## $ performance <dbl> 0.9003513, 0.9027556, 0.9055994, 0.9028525, 0.9097631, 0.9…Let’s visualise monthly four hour waiting time target performance before we save or raw data file.
ggplot2 is a powerful package to draw graphics. It implements the grammar of graphics (and hence its name). The ggplot2 package is loaded by the tidyverse package, as one of its core components. The ggplot() function initialises a ggplot object. We will use the geom_line() to connect and geom_point() to represent the percentage of accident and emergency attendances that met the four hour standard (y), ordered by month of attendance(x). We will use the labs() function to add the x and y labels, plot title and caption.
ggplot(ENG_performance, aes(period, performance)) +
geom_line(color = "darkcyan") +
geom_point(color = "darkcyan") +
scale_y_continuous(labels = percent) +
scale_x_date(date_labels = "%b-%y", date_breaks = "11 month")+
labs(x = "Month of attendance",
y = "% of A&E attendances that met the four hour standard",
title = "NHS England accident and emergency (A&E) four hour performance",
caption = "Source: NHSRdatasets")
The National Health Service (NHS) is always under considerable pressure over the winter period as demand for services tends to increase significantly with the onset of cold weather and flu.(NHS Providers, 2022) You can clearly see that under the winter pressures where four hour waiting time target performance drops.
Let’s select the ae_attendances data subset for further exploratory analysis
Based on this brief probe of the of the ae_attendances data, we have decide we wish to explore this data further for the development and evaluation your data capture tool.
Let’s tabulate the subsetted ae_attendances data for your report
ae %>%
# set the period column to show in Month-Year format
mutate_at(vars(period), format, "%b-%y") %>%
# set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches), comma) %>%
# show the first 10 rows
head(10) %>%
# format as a table
kable()| index | period | attendances | breaches |
|---|---|---|---|
| 1 | Mar-17 | 21,289.0 | 2,879.0 |
| 2 | Mar-17 | 813.0 | 22.0 |
| 3 | Mar-17 | 2,850.0 | 6.0 |
| 4 | Mar-17 | 30,210.0 | 5,902.0 |
| 5 | Mar-17 | 807.0 | 11.0 |
| 6 | Mar-17 | 11,352.0 | 136.0 |
| 7 | Mar-17 | 4,381.0 | 2.0 |
| 8 | Mar-17 | 19,562.0 | 258.0 |
| 9 | Mar-17 | 17,414.0 | 2,030.0 |
| 10 | Mar-17 | 7,817.0 | 86.0 |
Let’s save provisional subsetted ae_attendances data to the ‘RawData’ folder
Of note, when naming folders and files, it is important that you do so in a consistent, logical and predictable way means that information may be located, identified and retrieved by your and your colleagues, as quickly and easily as possible. With this in mind let’s name this file “ae_attendances_ENG_4hr_perfom”, and write it to the raw data folder.
glimpse(ae)## Rows: 12,765
## Columns: 4
## $ index <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,…
## $ period <date> 2017-03-01, 2017-03-01, 2017-03-01, 2017-03-01, 2017-03-0…
## $ attendances <dbl> 21289, 813, 2850, 30210, 807, 11352, 4381, 19562, 17414, 7…
## $ breaches <dbl> 2879, 22, 6, 5902, 11, 136, 2, 258, 2030, 86, 1322, 140, 0…write_csv(ae, here("RawData", "ae_attendances_ENG_4hr_perfom.csv"))Separating provisional ae_attendances_ENG_4hr_perfom data into training and testing sets
To develop and evaluate your data capture tool, you will need to splint the raw data into test and training data sets. We have added an index column earlier, so we can link the partitioned data sets to the raw ae_attendances and ae_attendances_ENG_4hr_perfom if required in the future.
How many rows are in the ae_attendances_ENG_4hr_perfom dataset?
#The ae_attendances_ENG_4hr_perfom dataset is large with
nrow(ae) #rows of data## [1] 12765We do not want you to spend hours inputting rows and rows data into your data capture tool. A test data set of 10-15 records to capture with and evaluate your data capture tool is sufficient for the purpose of the course assignment. Here is how to work out the proportion (prop) of the raw data to assign to the training data:
prop<-(1-(15/nrow(ae)))
#The proportion of the raw that needs to be assigned to the training data to ensure there is only 10 to 15 records in the test data is:
print(prop)## [1] 0.9988249We will use the createDataPartition() function from the caret package to splint our raw data into test and training data sets. The ‘set.seed()’ function is a random number generator, which is useful for creating random objects that can be reproduced. This will make sure that every time we run this script, we will partition the raw data into the same test and training data.
set.seed(333)
#Partitioning the raw data into the test and training data.
trainIndex <- createDataPartition(ae$index, p = prop,
list = FALSE,
times = 1)
head(trainIndex)## Resample1
## [1,] 1
## [2,] 2
## [3,] 3
## [4,] 4
## [5,] 5
## [6,] 6# All records that are in the trainIndex are assigned to the training data.
aeTrain <- ae[ trainIndex,]
nrow(aeTrain)## [1] 12753There are 12,753 records in your training data. That is a large dataset!
Let’s tabulate ae_attendances_ENG_4hr_perfom training data for your report
aeTrain %>%
# set the period column to show in Month-Year format
mutate_at(vars(period), format, "%b-%y") %>%
# set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches), comma) %>%
# show the first 10 rows
head(10) %>%
# format as a table
kable()| index | period | attendances | breaches |
|---|---|---|---|
| 1 | Mar-17 | 21,289.0 | 2,879.0 |
| 2 | Mar-17 | 813.0 | 22.0 |
| 3 | Mar-17 | 2,850.0 | 6.0 |
| 4 | Mar-17 | 30,210.0 | 5,902.0 |
| 5 | Mar-17 | 807.0 | 11.0 |
| 6 | Mar-17 | 11,352.0 | 136.0 |
| 7 | Mar-17 | 4,381.0 | 2.0 |
| 8 | Mar-17 | 19,562.0 | 258.0 |
| 9 | Mar-17 | 17,414.0 | 2,030.0 |
| 10 | Mar-17 | 7,817.0 | 86.0 |
Our next task, it to save ae_attendances_ENG_4hr_perfom training data to your working data folder ‘Data’
write_csv(aeTrain, here("Data", "ae_attendances_ENG_4hr_perfom_train.csv"))Let’s extract the ae_attendances_ENG_4hr_perfom test data
#All records that are not in the trainIndex (-trainIndex) are assigned to the test data.
aeTest <- ae[-trainIndex,]
nrow(aeTest)## [1] 12There are 12 records in your test data. Perfect! You now need to set aside the first record from the ae_attendances_ENG_4hr_perfom test data so that your markers can test and evaluate your data-capture tool.
aeTestMarker <- aeTest[1,]Let’s tabulate ae_attendances_ENG_4hr_perfom marker test data for your report
aeTestMarker %>%
# set the period column to show in Month-Year format
mutate_at(vars(period), format, "%b-%y") %>%
# set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches), comma) %>%
# show the first 10 rows
head(10) %>%
# format as a table
kable()| index | period | attendances | breaches |
|---|---|---|---|
| 115 | Mar-17 | 309 | 1 |
Our next task, it to save our ae_attendances_ENG_4hr_perfom marker test data to our working data folder ‘Data’
write_csv(aeTestMarker, here("Data", "ae_attendances_ENG_4hr_perfom_test_marker.csv"))We then need to set aside the remaining records for you to test (or collect with your) your data-capture tool.
aeTest <- aeTest[2:nrow(aeTest),]Let’s tabulate ae_attendances_ENG_4hr_perfom test data for your report
aeTest %>%
# set the period column to show in Month-Year format
mutate_at(vars(period), format, "%b-%y") %>%
# set the numeric columns to have a comma at the 1000's place
mutate_at(vars(attendances, breaches), comma) %>%
# show the first 10 rows
head(10) %>%
# format as a table
kable()| index | period | attendances | breaches |
|---|---|---|---|
| 1155 | Dec-16 | 200 | 0.0 |
| 2059 | Oct-16 | 6,452 | 360.0 |
| 3468 | May-16 | 417 | 0.0 |
| 4153 | Mar-18 | 9,376 | 112.0 |
| 4820 | Feb-18 | 245 | 0.0 |
| 7243 | Jul-17 | 5,170 | 235.0 |
| 8057 | Apr-17 | 15,957 | 1,309.0 |
| 8957 | Feb-19 | 7,258 | 1,374.0 |
| 10214 | Oct-18 | 3,197 | 0.0 |
| 10328 | Oct-18 | 2,033 | 8.0 |
Our final task, is to save our ae_attendances_ENG_4hr_perfom test data to our working data folder ‘Data’
write_csv(aeTest, here("Data", "ae_attendances_test.csv"))You are now ready to start exploring your data. Happy coding!